Scientists have found new autism genes by scanning the largest collection of families with multiple cases of autism ever assembled.
Chromosome 11 was identified as one of the culprits
The monumental task of studying the 1,200 families took more than 120 scientists from more than 50 institutions across 19 countries.
The work, described in Nature Genetics, implicates a region of chromosome 11 and a specific gene called neurexin 1.
Experts say the findings should help with finding new autism treatments.
Autism is a complex brain disorder that inhibits a person's ability to communicate and develop social relationships, and is often accompanied by extreme behavioural challenges.
Scientists have long suspected that errors in a person's genetic blueprint play a part in autism.
Genetic code, DNA, is packaged in cells in the form of chromosomes, which are made up of strings of genes.
Experts have now located a region of one chromosome - 11p - and a gene called neurexin 1.
Neurexin belongs to a family of genes that help nerve cells communicate and the scientists now believe these play a critical role in autism spectrum disorders.
The work began five years ago in 2002, when scientists pooled their research resources and expertise and formed the Autism Genome Project.
They used "gene chip" technology to look for genetic similarities in autistic people, funded by the non-profit organisation Autism Speaks and the US National Institutes of Health.
Child psychiatrist Professor Jonathan Green, who led the clinical fieldwork in Manchester, said: "Autism is a very difficult condition for families - communication is taken for granted by parents of healthy children but is so greatly missed by those with autistic children.
"We hope that these exciting results may represent a step on the way to further new treatments in the future."
Professor John Burn, of the Institute of Human Genetics at Newcastle University, said the news was fantastic.
But he cautioned: "There will almost certainly be an interaction between several genes so this one discovery doesn't provide a complete answer and may not lead straight to a genetic test but it could be a key step in development for effective treatments as it provides a target for drug development."
Dr Fred Kavalier of the British Society for Human Genetics said a cure for autism was still a long way off.
Professor Simon Baron-Cohen, director of the Autism Research Centre at Cambridge University, said independent replication of the findings was now needed.
The next phase of the work, which will take three years and cost more than £7m, will home in on the suspect areas of DNA.
It is estimated that autism and related disorders affect up to one in 100 British children.